[Adapt] [Seminar]From Recognition to Cognition: Visual Commonsense Reasoning

[周泽林]

Hi Adapters,




I'm Zelin. In this seminar, I'll introduce you with a 2019 CVPR paper "From Recognition to Cognition: Visual Commonsense Reasoning", which constructed a popular commonsense dataset VCR and presented a multimodal reasoning engine. As the combination of NLP and CV arousing much concern recently, commonsense will have broader applications in visual domain. 




url: https://visualcommonsense.com




Time: Wed 4:00pm

Venue: SEIEE 3-414

 

Best Regards,

Zelin Zhou
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.sjtu.edu.cn/pipermail/adapt/attachments/20200929/edea332a/attachment-0001.html>